The RATS Software Forum

Good morning

I have written my own code for the KSS (Kapetanios, Shinb and Snell, 2003) test and seems to work well. Nevertheless, for my time series under study the values for the t-statistic are usually positive (like 80% of the cases). This is technically correct according to the t-Student distribution, but I would like to know if there could be any mistake or something that I am not taking into account.

Thank you beforehand.

It's a t-statistic but it's not distributed at all like a t. (That's why there are look-up tables). Under the null, about 80% of the values are negative, so 80% positive would be rather unlikely.